-C

hri hiao ia S d fo ne 1 by f la al clattices, and the characteristic path lengths increase logarithmically with the number of nodes. One of the elementary moves, rigid rotation, has drastic effect on the geometric properties of the network. The move increases greatly the connections and reduces significantly the shortest path lengths between conformations. Including rigid rotation to the move set results in the increase of the dimensionality of the conformation space to the value about 4. © 2006 Elsevier B.V. All rights reserved. Protein folding is a complex process for which, a sequence of amino acids folds into a unique and stable structure in a relatively short time [1]. The lattice models have been used widely as coarse-grained models for the theoretical study of folding process [2–7]. In the lattice models, protein is viewed as a chain of m monomers, and the conformations are given by all possible self-avoiding walks of the chain on a two or three-dimensional lattice. The energy of a conformation, in gen-eral, depends on the number of intrachain contacts, and how to assign the contact energy is model dependent. The kinetics of folding process can then be studied by Monte Carlo sim-ulations for which, a move set is designed for the change of conformations. In principle, different move sets, satisfying the requirement of ergodicity, should reach the same equilibrium canonical distribution after sufficiently long time simulations. However, different move sets may yield different perspectives of folding kinetics. Chan and Dill analyzed the folding kinetics of two different move sets for 2D homo- and hetero-polymers by using the Metropolis transfer matrix method [5,6]. Their * Corresponding author